Child Neglect, Social Context, and Educational Outcomes: Examining the Moderating Effects of School and Neighborhood Context
Chapple, Constance L., Vaske, Jamie, Violence and Victims
Research on child neglect has found that neglected children are more likely to experience worse developmental outcomes than non-neglected children. These negative outcomes include antisocial behavior as well as poor school performance. Eco-developmental theory has found that adverse social contexts often worsen these outcomes for neglected and maltreated youths. However, little research has been done on the educational outcomes of neglected children and none of it has employed a national, longitudinal, community sample with an examination of social context. We do so in our research and find that several types of child neglect significantly predict a variety of poor educational outcomes at the bivariate level and that physical and educational neglect were significantly associated with a composite measure of school problems in multivariate analysis. We offer several explanations for our findings and future directions for research.
Keywords: child neglect; educational outcomes; community organization; school organization; eco-developmental theory
Each year, nearly half a million children in the United States are victims of child neglect (Sedlak & Broadhurst, 1996 ). Neglected children often face severe developmental challenges such as poor school performance (Eckenrode, Laird, & Doris, 1993; Leiter & Johnsen, 1997; Thornberry, Ireland, & Smith, 2001 ) and antisocial behavior (Ireland, Smith, & Thornberry, 2002; Zingraff, Leiter, Myers, & Johnsen, 1993 ). Clearly, neglected children are not confined to families alone but also belong to communities and attend schools. However, few studies of child neglect or child maltreatment in general have examined the moderating role that social context plays in the developmental outcomes of neglected children (Zielinski & Bradshaw, 2006 ).
Research indicates that the ill effects of child neglect are pernicious (Chapple, Tyler, & Bersani, 2005 ) and that child neglect, unlike physical abuse, is most likely a deficit of early childhood, which can produce serious developmental deficits. Current studies of child neglect have focused on individual processes, such as emotional regulation, attachment disorders, and association with deviant peers (Bolger & Patterson, 2001; Bolger, Patterson, & Kupersmidt, 1998; Chapple et al., 2005; De Paul & Arruabarrena, 1995; Herrenkohl, Huang, Tajima, & Whitney, 2003; Maughan & Cicchetti, 2002 ) while the social context of child neglect has rarely been investigated (see Coulton, Korbin, & Su, 1999; Garbarino & Sherman, 1980 for exceptions). The research that has examined the social context of child neglect has done so with global measures of child maltreatment, leaving unanswered the question of whether child neglect is uniquely influenced by social context.
We know that young, disadvantaged mothers are most likely to neglect their children (Coulton et al., 1999 ), but we have little information on the community and school contexts in which neglected children live and go to school. Prior research on child maltreatment using eco-developmental theory has focused primarily on antisocial behavior, with relatively little attention to educational outcomes (Coulton et al., 1999; Garbarino & Sherman, 1980; Stouthamer-Loeber, Wei, Homish, & Loeber, 2002 ). However, as researchers acknowledge, educational difficulties often have long-term repercussions into adulthood. According to ecological theory, children are nested within micro, meso, and macro contexts that exert independent and multiplicative effects on development (Bronfenbrenner, 1977; Garbarino, 1992). These contexts have both proximal and distal effects (Zielinski & Bradshaw, 2006 ) on children's behavior and cross-level interactions are assumed. In such a scenario, school and community organization may act as either a risk or a protective factor for neglected children's educational outcomes. We address the question of whether social context moderates the effects of child neglect on children's educational outcomes with a prospective, longitudinal, community sample of children and their mothers. First, we turn to a review of the literature.
Child neglect, unlike other types of maltreatment, is largely confined to early childhood (Connell-Carrick & Scannapieco, 2006 ). As such, researchers are faced with a set of unique methodological problems and potential policy benefits. Because the effects of child neglect occur early in the life-course, yet may be felt for years to come, researchers need longterm longitudinal data. Also, the great majority of research on child neglect has used official records or has followed children who had officially recognized cases of child neglect (Gershater-Molko, Lutzker, & Sherman, 2003; Hildyard & Wolfe, 2002 ). While this is standard research practice on a hard-to-survey population, child advocates acknowledge that often only the most extreme cases of child neglect are referred to social services. Children who are the victims of less serious but still deleterious neglect are often missed in these samples (Chapple et al., 2005 ). Therefore, recent reviews have called for the use of longitudinal data from community samples to examine the wide-ranging levels of neglect and the negative long-term consequences of less severe cases of maltreatment (Zielinski & Bradshaw, 2006 ). The benefit, however, is that because child neglect is often confined to early childhood (before age 5), there are many ameliorative variables to investigate that occur before adverse outcomes emerge but after the neglect has occurred. Social context, in the form of community and educational organization, and climate are two such variables.
Child maltreatment researchers have long noted that types of maltreatment overlap (e.g., physically abused children are also neglected) (Understanding Child Abuse and Neglect, 1993). Research on child neglect is relatively recent and as such, we often refer to the general child maltreatment literature and note specific research on child neglect where available. Child maltreatment interferes with development and has been tied to a host of adverse outcomes such as delinquency (Ireland et al., 2002; Zingraff et al., 1993 ) and poor school performance (Eckenrode et al., 1993; Leiter & Johnsen, 1997; Thornberry et al., 2001 ). At the same time, very little research has assessed the independent effect of different types of maltreatment on developmental outcomes. The few pieces that do, suggest that neglected children suffer similar (Leiter & Johnsen, 1997; Lemmon, 1999 ) or greater adverse developmental outcomes as physically abused children (Eckenrode et al., 1993; Zingraff et al., 1993 ). Because the research on child neglect is limited, we review research on child maltreatment and educational outcomes, and on maltreatment and social organization later.
Child Maltreatment and Academic Outcomes. Although much research in this area has suffered methodological problems such as small sample sizes and the lack of a comparison group, research indicates that maltreated children often do more poorly in school than nonmaltreated children. Abused and/or neglected children and youths have lower IQ (Veltman & Browne, 2001), lower grades (Eckenrode, Rowe, Laird, & Brathwaite, 1995; Kendall- Tackett & Eckenrode, 1996; Leiter & Johnsen, 1994, 1997 ), poorer reading and other test scores (Eckenrode et al., 1995; Kinard, 2001a), lower school expectations, higher grade repetitions or drop-out (Eckenrode et al., 1993; Thornberry et al., 2001 ), and greater school disciplinary problems (Kendall-Tackett & Eckenrode, 1996; Leiter & Johnsen, 1997 ) than nonmaltreated youths. Research has also indicated that neglected children do as poorly as abused children (Leiter & Johnsen, 1997 ) or have worse academic outcomes (Kendall- Tackett & Eckenrode, 1996; Kinard, 2001a, 2001b). Research indicates that chronically maltreated children or maltreatment in late childhood and into adolescence has a particularly adverse effect on school adjustment (Thornberry et al., 2001 ).
Child Maltreatment, Social Organization and Development. Past research on child development and community organization has found that children from efficacious communities have more positive developmental outcomes (Browning & Olinger-Wilbon, 2003; Rankin & Quane, 2002 ). However, communities evolve over time and children and adolescents affect communities just as communities affect them. Therefore, we need longitudinal examinations during all points in the life-course to understand how child development and community organization interact. Additionally, recent articles have called for research on child maltreatment that utilizes prospective, longitudinal data into adolescence from community samples (Gershater-Molko et al., 2003; Hildyard & Wolfe, 2002 ). Our research addresses both of these needs as we examine the moderating effects of social context and organization over time for neglected children in a community sample.
Child maltreatment researchers utilizing an eco-developmental theoretical framework have long acknowledged that social context influences child maltreatment (Coulton et al., 1999; Garbarino & Sherman, 1980; Stouthamer-Loeber, Wei, Homish, & Loeber, 2002 ). According to research on the predictors of child maltreatment, maltreatment is more likely to occur in neighborhoods with high levels of crime or social disorganization (Coulton, Korbin, & Su, 1996; Garbarino & Crouter, 1978 ). In particular, maltreating mothers are more likely than non-maltreating mothers to indicate that their community is unsafe or disorganized (Korbin, 2003; Polansky, Gaudin, Ammons, & Davis, 1985 ). Also, research has found that maltreating families are more socially isolated (Vinson, Baldry, & Hargreaves, 1996 ). These families could be the ideal benefactors of efficacious communities and schools, which could aid a neglected child's development.
Eco-developmental researchers in child maltreatment have moved away from macrolevel measures of social structure and social (dis)organization and toward concepts such as social cohesion or collective efficacy to measure not just how communities are structured but, rather, how they function (Korbin, 2003 ). Community collective efficacy, a term developed by Sampson, Raudenbush, and Earls (1997), measures community organization, community social control, cohesion, and trust among community residents. Researchers have theorized that in efficacious communities, "many hands make light work" because the socialization and supervision of children and youths in the community lies with the community and not with individual families alone (Shaw & McKay, 1942 ). In such communities, trust and cooperation predominate and children and families experience the "strength of weak ties" (Granovetter, 1973 ) in which community residents are linked via strong, yet distant ties. Few studies of child maltreatment and child and adolescent development have investigated the moderating effect of community processes such as community and school organization, although preliminary research indicates that neglected children from disorganized communities suffer particularly poor outcomes (Coulton et al., 1999; see Crooks, Scott, Wolfe, Chiodo, & Killip, 2007, for a notable exception).
Added into this mix of research on child maltreatment and community organization is the concept of educational organization and/or climate. Maltreatment researchers have found that school climate plays an important part in ameliorating the effects of maltreatment on adverse outcomes (Crooks et al., 2007 ). In their research, Crooks et al. (2007) found that school climate, measured by perceptions of safety and the availability of a violence prevention program, reduced the likelihood of violence for maltreated youths. Researchers also have found that schools that exhibit collective efficacy have higher student achievement in general (Brookmeyer, Fanti, & Henrich, 2006 ). Efficacious schools, as well as efficacious communities, may be even more critical in the development of neglected children (Heller, Larrieu, D'Imperio, & Boris, 1999 ) than in non-maltreated children as these children have few social supports. Our research addresses several issues in the literature. First, per recent recommendations, we analyze data from a longitudinal community sample that tracks child development from birth into adulthood. This data allows us to investigate the long-term effects of neglect for children who are not necessarily officially recognized neglect cases. This data also contains a large comparison group of children who have not been neglected. Second, because educational outcomes for neglected children is an under-researched topic and early educational problems often create long-term deficits, we investigate the effect of neglect on grade retention, school problems, suspension, and remedial classes in early adolescence. Finally, our analysis of the moderating effect of neglect on educational outcomes provides important insight into the social context of child neglect and potential ameliorating processes. We propose the following hypotheses: first, we suggest that child neglect will be associated with greater adverse educational outcomes such as school behavioral problems, attending remedial classes, suspension and/or expulsion, and grade retention. Second, we propose that weaker community and school organization will amplify the negative effect of child neglect on school outcomes so that neglected children in poorly organized communities and schools will have the worst educational outcomes.
DATA AND METHODS
We use data from the NLSY79 Child survey to examine whether neighborhood and school organization moderate the effects of child neglect on a host of educational outcomes. The original National Longitudinal Survey of Youth (NLSY79) was a nationally representative longitudinal sample of youth aged 14 to 21 in 1979. The NLSY79 contains "extensive information about the employment, education, training, and family experiences of the respondents" (Center for Human Resource Research, 2000, p. 1). Beginning in 1982, the female respondents of the NLSY were surveyed about "pregnancy, postnatal fertility, and child care experiences" (Center for Human Resource Research, 2000, p. 1). Since 1986, assessments of the children born to the women of the NLSY have been conducted. The NLSY79 Child and Young Adult sample "approximately represent[s] a cross-section of children born to a nationally representative sample of women who were between the ages of 31 and 38 on January 1, 1998" (Center for Human Resource Research, 2000, p. 3). The NLSY79-Child and Young Adult data has been used extensively in educational, developmental, and delinquency research.
We combined data from mother-reported supplements and interviewer's observations in 1988, when children were between the ages of 2 and 5 years old, with mother-reported data on educational outcomes from 1996. Youths were between the ages of 10 and 13 years old in 1996. The measures of neglect were collected in 1988, and the neighborhood and school organization variables were measured in 1996 (coterminous with the outcomes). All dependent variables (i.e., school problems, remedial classes, suspension, and grade retention) were measured in 1996. The final analytical sample included data from 1,080 subjects.
A number of measures were missing data from respondents. Similar to previous analyses of the NLSY (Chapple et al., 2005 ), we found that cases that were missing data on at least one variable were more likely to be "at-risk" youths. Cases that were missing data were more likely to be non-White, report higher levels of physical and emotional neglect, report lower levels of neighborhood and educational organization, and were more likely to be suspended. Cases missing data, however, were less likely to retain a grade between kindergarten and 9th grade. In light of these differences, caution should be used when interpreting the current study's findings.
We examine the independent and interactive effects of educational neglect, physical neglect, emotional neglect, neighborhood organization, and school organization on educational outcomes, controlling for the effects of gender, race, and poverty.
The neglect measures were taken from the HOME-SF (short form). 1 Educational neglect was created from two items that inquired about the mothers' involvement in the child's learning activities. More specifically, mothers were asked whether they helped their child learn the alphabet, numbers, colors, and shapes (0 = no, 1 = yes ). We summed these answers to form a learning variable (0 = none to 4 = all four ). Mothers were also asked how often they read to their child (0 = never, 1 = several times a year, 2 = several times a month, 3 = once a month, 4 = about 5 times a week, 5 = every day ). These two items were summed to form a composite scale of educational neglect. The scale was formed so that lower values meant higher levels of neglect. The mean of the scale was 6.75, which indicated low levels of educational neglect (Table 1 ).
We created a measure of physical neglect from three variables in which the interviewers rated whether the home appeared safe, reasonably clean, and minimally cluttered. The three dichotomous items (0 = no, 1 = yes ) were summed to form a composite scale of physical neglect, so that lower scores reflected higher levels of neglect (α =.73).
Emotional neglect is an interviewer-reported observation of the mother's verbal interactions with her child during the 1988 interview. Interviewers reported whether the mother "conversed with the child at least twice during the interview," whether the mother "answered the child's questions verbally," and whether the mother "spoke to the child at least twice during the interview." These questions were coded as 0 = no and 1 = yes. The three items were summed to form a composite index of emotional neglect, which was coded so that lower scores indicated neglect (α =.74). The scale's scores ranged from 0 to 3, with a scale mean of 2.69. This indicated that, on average, youths experienced low levels of emotional neglect. Previous researchers who have used the NLSY have used similar measures of educational, physical, and emotional neglect (Chapple et al., 2005 ).
The neighborhood organization and climate variable was composed of eight items reflecting mothers' perceptions of their neighborhood. Mothers were asked to rate how big of a problem various issues were within their neighborhood (1 = a big problem, 2 = somewhat of a problem, and 3 = not a problem ). More specifically, mothers rated the following eight statements: people don't respect rules and laws, crime and violence, abandoned or run-down buildings, not enough police protection, not enough public transportation, too many parents don't supervise their children, people don't care what goes on in the neighborhood, and lots of people cannot find jobs. This measure captures some of the concepts relevant to neighborhood organization, such as social cohesion, access to institutional resources, and informal and formal social control (Sampson, Morenoff, & Gannon-Rowley, 2002 ). Further, previous researchers who have used the NLSY79-Child data have employed similar measures to measure neighborhood organization and quality (Colder et al., 2006 ). The eight items were coded so that lower scores indicated weak neighborhood organization (α =.84).
The school organization and climate measure was created using eight items that assessed mothers' perception of their child's school (see Brookmeyer, Fanti, & Henrich, 2006, for a similar measure). Mothers were asked to rate their child's school (1 = fail, 2 = D, 3 = C, 4 = B, and 5 = A ) on how well the "teachers are caring," "principal is a leader," "teacher's skills," "safety at school," "letting parents know students' progress," "letting parents help make decisions," "teachers teaching students right from wrong," and "school maintains order." The scores from these eight items were summed to form a composite measure of school organization. The variable was coded so that lower scores reflected weaker school organization (α =.91). The scale's mean was 32.18 and the median was 33, indicating that a substantial percentage of mothers perceived that their child's school was moderately organized.
We also control for the effects of gender, race, and family's poverty status in our analysis. Gender was coded as 0 = female and 1 = male. Race is coded as 0 = White and 1 = non- White. Approximately 51% were White, 29% were Black or African American, and 20% were Hispanic. Youths' family poverty status was assessed between the years of 1985 and 1990. Responses were coded dichotomously (0 = never lived in poverty any years between 1985 and 1990, and 1 = living in poverty during any of these years ).
We examine the effects of neglect and social organization on four mother-reported dependent variables: school problems, remedial classes, suspension, and grade retention. First, mothers were asked whether their child had "any behavior problems at school resulting in your receiving a note or being asked to come in and talk to the teacher or principal" (0 = no, 1 = yes ). Second, mothers were also asked, "Does your child go to a special class or get special help in school for remedial work?" Responses were coded as 0 = no and 1 = yes as this was the metric in which the questions were originally scored. Third, mothers were asked whether their "child has ever been suspended or expelled from school" (0 = no, 1 = yes ). Finally, mothers reported whether their student had been held back in any grade from kindergarten to 9th grade (0 = no, 1 = yes ). It is important to note that the dependent variables capture the prevalence of behavioral and educational problems, rather than the frequency and variety of problems. The variables are measured dichotomously because this is how the measures were originally scaled by the NLSY researchers. We also created a composite variable, total educational problems, to see if neglect was associated with a range of negative educational outcomes. Each of the individual dependent variables were added together to create this variable that ranges in score from 0 to 4. High scores indicate that the child experienced many, if not all, of the previously mentioned educational problems.
The current analyses are divided into three stages. First, we assess the bivariate relationships between the neglect measures, neighborhood organization, school organization, and the four outcome variables. Second, we use logistic regression to estimate the effects of educational neglect, physical neglect, and emotional neglect on the five educational outcomes (school problems, remedial classes, suspended, grade retention, and total educational problems), controlling for the effects of gender, race, and poverty. Finally, we use logistic regression to estimate the independent and interactive effects of the three measures of neglect, neighborhood organization, and school organization on the four dependent variables, controlling for the effects of gender, race, and poverty. We use negative binomial regression to estimate the total educational problems model as it is an overdispersed count variable. The interaction terms were created by multiplying a neglect measure by either neighborhood organization or school organization. The three neglect measures, neighborhood organization, and school organization variables were mean-centered prior to creating the interaction variables in order to reduce multicollinearity between the variables that constitute the interaction variable (Aiken & West, 1991 ). In an effort to conserve space, we have only presented results for the significant interaction effects.
The regression models were estimated using STATA 9.0. Regression coefficients are interpreted as the amount of change in the logit or logged odds [ln(P i / 1 - P i )] that occurs when the independent variable changes by 1 unit. However, logits are difficult to interpret, and researchers typically exponentiate the coefficient to produce the odds ratio (OR). The OR is interpreted as the odds of an event occurring for one group or category of X compared to the odds of the event occurring for another group. The formula, [(exp(b)-1)*100], can also be used to get the percentage of change in the dependent variable that occurs when the independent variable increases by 1 unit. Models include estimates of Huber/ White corrected standard errors to account for the clustering of siblings within households (Rogers, 1993 ), because the NLSY includes data from multiple children within a household. Models that do not correct for the clustering or nesting of observations may produce biased estimates of the standard error, which may produce misleading coefficient tests.
Descriptive statistics are shown in Table 1. As shown in the table, 49% of the sample are males, 48% are non-White, and 46% of mothers reported that they lived in poverty between 1985 and 1990. In addition, the descriptive statistics suggest that youths, on average, experience low levels of educational neglect, physical neglect, and emotional neglect, and relatively high levels of neighborhood organization and educational organization. Approximately 33% of mothers reported that their child had behavior problems in school in 1996. Fifteen percent of the youths attended a special or remedial class, 16% of youths were suspended or expelled from school in 1996, and 15% of youths had been retained in at least one grade between kindergarten and 9th grade.
The first step in the analyses is to examine the bivariate relationships between the endogenous and exogenous variables. Analyses reveal that higher levels of educational neglect are significantly related to a higher prevalence of behavioral problems in school (χ^sup 2^ = 16.21, p =.039), taking remedial classes (χ^sup 2^ = 16.34, p =.038), suspension (χ^sup 2^ = 32.38, p <.001), grade retention (χ^sup 2^ = 22.86, p =.004), and weaker neighborhood organization (F = 4.42, p <.001). Higher levels of physical neglect correspond to weaker neighborhood organization (F = 5.05, p =.002) and a higher likelihood of being placed in remedial classes (χ^sup 2^ = 7.92, p =.048). Greater emotional neglect is related to a higher prevalence of suspension (χ^sup 2^ = 14.43, p =.002), a higher prevalence of grade retention (χ^sup 2^ = 8.36, p =.039), and weaker neighborhood organization (F = 6.66, p <.001). Emotional neglect is significantly related to educational neglect (r =.126, p <.001) and physical neglect (r =.128, p <.001), and physical neglect is associated with educational neglect (r =.080, p =.016). T -test analyses reveal that youths who have school problems (t = 2.10, p =.038), who take remedial classes (t = 2.02, p =.044), who have been suspended (t = 4.00, p <.001), and who have retained a grade (t = 3.62, p <.001) are more likely to come from poorly organized neighborhoods than youths who have not had school problems, not taken remedial classes, not been suspended, and not retained a grade. Finally, youths who have school problems (t = 4.48, p <.001) and who have been suspended (t = 3.32, p <.001) are more likely to report weaker educational organization than youths who have not had school problems and who have not been suspended.
The results from the logistic and negative binomial regression models are displayed in Table 2. As shown in the table, gender and school organization are significantly related to school behavioral problems. The odds of having behavioral problems in school are nearly three times higher for males than females. Exponentiation of the school organization coefficient reveals that each unit increase in school organization decreases the odds of having a behavioral school problem by 4.1%. Level of neighborhood organization, educational neglect, and emotional neglect are not significantly related to school problems. Physical neglect has a marginally significant effect on the likelihood of school problems. The negative coefficient suggests that the odds of school problems are approximately 20% lower among youths who experience little to no physical neglect.
Column 3 of the table shows that the likelihood of taking remedial classes is significantly lower among those who are not educationally neglected; however, the coefficient becomes nonsignificant when neighborhood organization and school organization are added to the model. Gender has a positive effect on the odds of taking remedial classes. More specifically, the odds of taking remedial classes are 1.7 times greater for males compared to females.
The first estimates for the suspension model show that gender, race, and poverty are positively related to suspension. More specifically, the odds of suspension are approximately two times higher among males, three times higher among non-Whites, and two and a half times higher among impoverished youths than non-impoverished youths. Also, there is a significant interaction between neighborhood organization and emotional neglect on the likelihood of suspension. As shown in Figure 1, neighborhood organization influences the likelihood of suspension for youths who have moderate levels of emotional neglect (i.e., emotional neglect = 1 and emotional neglect = 2). This suggests that likelihood of suspension is higher among youths who experience moderate levels of emotional neglect and who live in well-organized neighborhoods, compared to youths who experience moderate levels of emotional neglect and who live in poorly organized neighborhoods. The figure also shows that the probability of suspension does not vary by level of neighborhood organization when youths experience very low levels of emotional neglect (i.e., emotional neglect = 3) or when they experience very high levels of emotional neglect (i.e., emotional neglect = 0). In fact, youths who experience high levels of emotional neglect (i.e., emotional neglect = 0) and who come from poorly organized neighborhoods (i.e., -1 SD below the mean) have approximately the same probability of suspension as youths who experience high levels of emotional neglect and who come from well-organized neighborhoods (i.e., +1 SD above the mean). The same pattern of results is found among youths who experience very low levels of emotional neglect. 2
The results for the grade retention model are shown in the last two columns of Table 2. The results show that the likelihood of being held back a grade is higher among males, non- Whites, and impoverished youths. The probability of retaining a grade is 18% for males, compared to 11% for females. In addition, the odds of retaining a grade are approximately two times higher for non-White youths than White youths. Finally, nearly 3 out of every 10 impoverished youths retained a grade during elementary school, compared to 1 out of every 10 non-impoverished youths.
Finally, in the last column of Table 2, negative binomial regression results reveal that educational neglect and physical neglect are significantly associated with the composite measure of total educational problems. Boys, minority children, and children who lived in poverty were also more likely to have greater total educational problems. Children who experienced educational and/or physical neglect were more likely to report greater numbers of educational problems. These coefficients remain significant after controlling for neighborhood organization and educational organization. Further, the interaction between neighborhood organization and emotional neglect remains significant and is presented in Figure 2. Results suggest, much like the significant interaction for grade suspension, that neighborhood organization does not moderate the effect of child neglect on the total educational problems a child has for emotionally neglected children, but it does moderate the relationship for children who are not emotionally neglected.
DISCUSSION AND CONCLUSION
Our bivariate results showed that regardless of which outcome we evaluated, neglected children had far worse educational outcomes than non-neglected children. And child neglect in all its forms was related to worse school and community organization and climate at the bivariate level. However, our multivariate results for the individual school outcome measures suggest that educational outcomes for adolescents are affected much more strongly by demographic factors (race, age, poverty) than familial ones (neglect). However, when we examined the composite variable measuring a variety of negative school outcomes, both educational and physical neglect were associated with more school problems. This suggests to us that there may be a type of threshold effect that neglect has on school problems. When examining each individual outcome, neglected children are no more likely than non-neglected children to be suspended, report in-school behavioral problems, be retained a grade, or take remedial classes. However, when our outcome is a constellation of these negative outcomes, two measures of neglect are significant predictors even net of the strong effects of the demographic factors mentioned previously. This may mean that child neglect is a particular risk factor for high levels of educational failure. This finding should be followed up with other data from other samples to validate our supposition.
The results concerning our individual outcomes, although clear, might be an artifact of the data that we employ. The NSLY79 Child and Young Adult data oversamples non-White and poor respondents and includes an oversample of young mothers in the original cohort. As indicated in Table 1, nearly half of our respondents indicated that they lived in poverty (per income guidelines given to them by researchers) at some point during the study period. The ill effects of poverty on child and adolescent development clearly are confounded with child neglect and may have affected our results (null findings regarding child neglect on the individual outcome measures). As each group listed previously is more likely to have higher rates of child neglect than the general population, we may have too little variation in our sample to find important causal relationships between neglect and educational outcomes. Also, our measures of neglect, although used prior in community samples (Chapple et al., 2005 ), tap into relatively low to moderate levels of neglect. Perhaps there is a threshold effect of neglect on educational outcomes as we suggested in explanation of the results of our composite educational problems outcome. As compulsory education mandates and tracks students' attendance after age 5, perhaps educational outcomes are the last to be affected by dysfunctional family environments. Also, because there are so many safety nets in place for children who attend school, child neglect may only influence the most serious constellations of negative educational outcomes.
Also, the results we found regarding the significant interactions of emotional neglect and community organization on suspension and total educational problems are not without precedent. Wikstrom and Loeber (2000) reported a similar finding in which children at risk for delinquency were unaffected by neighborhood context, whereas in some instances in their research, at-risk children in moderately organized communities had higher levels of offending. This might also be true in our data as well. Although it is counter intuitive to think that neglected children in better social contexts would have worse developmental outcomes, perhaps these children, because they are an anomaly, fall through the cracks and go undetected. We believe this finding is worthy of future study and explanation. Our data has implications for both maltreatment researchers and practitioners. Our results suggest that child neglect does have an adverse effect on educational outcomes when they are viewed as a whole. Neglected children may not stand out when individual outcomes are assessed but rather child neglect becomes a significant risk factor when multiple negative outcomes are assessed. This might suggest that interventions for neglected children could ameliorate co-occuring academic deficiencies.
Our data clearly has some limitations with respect to the sampling of seriously neglected children. Obviously, it is unlikely that a mother who is seriously neglecting her children would participate in a survey, let alone participate biannually. However, other researchers using the same data (Chapple et al., 2005 ) found that even within this context, child neglect had pervasive negative effects on violent delinquency well into adolescence. Although our data is a limitation in this study, the longitudinal nature of the data is also a strength of our research. There are very few longitudinal community samples of youths that contain measures of child neglect in addition to a variety of etiological variables. And, to our knowledge, no national, longitudinal data on youths contains neglect measures. We would like to see additional work done on the outcomes of child neglect beyond antisocial and delinquent behavior. We believe that our work raises several important questions regarding the mechanisms in which neglected children interact within communities and schools.
1. The measurement of the neglect items was replicated from Chapple et al., 2005.
2. We also explored whether demographics moderated the effects of neglect on the four outcome variables. We did not find evidence that gender, race, or poverty moderated the effects of educational neglect, physical neglect, or emotional neglect on the likelihood of suspension, school behavioral problems, remedial classes, or grade retention.
Aiken, L., & West, S. G. (1991). Multiple regression: Testing and interpreting interactions. Newbury Park, CA: Sage Publications.
Bolger, K. E., & Patterson, C. J. (2001). Developmental pathways from child maltreatment to peer rejection. Child Development, 72, 549 - 568.
Bolger, K. E., Patterson, C. J., & Kupersmidt, J. B. (1998). Peer relationships and self-esteem among children who have been maltreated. Child Development, 69, 1171 - 1197.
Bronfenbrenner, U. (1977). Toward an experimental ecology of human development. American Psychologist, 32, 513 - 530.
Brookmeyer, K. A., Fanti, K. A., & Henrich, C. C. (2006). Schools, parents, and youth violence: A multilevel, ecological analysis. Journal of Clinical Child and Adolescent Psychology, 35(4), 504 - 514.
Browning, C. R., & Olinger-Wilbon, M. (2003). Neighborhood structure, social organization, and number of short-term sexual partnerships. Journal of Marriage and Family, 65, 730 - 745.
Chapple, C. L., Tyler, K. A., & Bersani, B. E. (2005 ). Child neglect and adolescent violence: Examining the effects of self-control and peer rejection. Violence and Victims, 20, 39 - 53.
Colder, C. R., Lengua, L. J., Fite, P. J., Mott, J. A., & Bush, N. R. (2006). Temperament in context: Infant temperament moderates the relationship between perceived neighborhood quality and behavior problems. Journal of Applied Developmental Psychology, 27, 456 - 467.
Connell-Carrick, K., & Scannapieco, M. (2006). Ecological correlates of neglect infants and toddlers. Journal of Interpersonal Violence, 21, 299 - 316.
Coulton, C. J., Korbin, J. E., & Su, M. (1996). Measuring neighborhood context for young children in an urban area. American Journal of Community Psychology, 24, 5 - 32.
Coulton, C. J., Korbin, J. E., & Su, M. (1999). Neighborhoods and child maltreatment: A multi-level study. Child Abuse and Neglect, 23, 1019 - 1040.
Crooks, C. V., Scott, K. L., Wolfe, D. A., Chiodo, D., & Killip, S. (2007 ). Understanding the link between childhood maltreatment and violent delinquency: What do schools have to add? Child Maltreatment, 12, 269 - 280.
De Paul, J., & Arruabarrena, M. I. (1995). Behavior problems in school-aged physically abused and neglected children in Spain. Child Abuse and Neglect, 19, 409 - 418.
Eckenrode, J., Rowe, E., Laird, M., & Brathwaite, J. (1995). Mobility as a mediator of the effects of child maltreatment on academic performance. Child Development, 66, 1130 - 1142.
Eckenrode, J., Laird, M., & Doris, J. (1993). School performance and disciplinary problems among abused and neglected children. Developmental Psychology, 29, 53 - 62.
Garbarino, J. (1992). Children and families in the social environment (2nd ed.). New York: Aldine De Gruyter.
Garbarino, J., & Sherman, D. (1980). High-risk neighborhoods and high-risk families: The human ecology of child maltreatment. Child Development, 51, 188 - 198.
Garbarino, J., & Crouter, A. (1978). Defining the community context for parent-child relations: The correlates of child maltreatment. Child Development, 49, 604 - 616.
Gershater-Molko, R. M., Lutzker, J. R., & Sherman, J. A. (2003). Assessing child neglect. Aggression and Violent Behavior, 8, 563 - 585.
Granovetter, M. S. (1973). The strength of weak ties. American Journal of Sociology, 78, 1360 - 1380.
Heller, S. S., Larrieu, J. A., D'Imperio, R., & Boris, N. W. (1999). Research on resilience to child maltreatment. Child Abuse and Neglect, 23, 321 - 338.
Herrenkohl, T. I., Huang, B., Tajima, E. A., & Whitney, S. D. (2003). Examining the link between child abuse and youth violence. Journal of Interpersonal Violence, 18, 1189 - 1208.
Hildyard, K. L., & Wolfe, D. A. (2002). Child neglect: Developmental issues and outcomes. Child Abuse and Neglect, 26, 679 - 695.
Ireland, T. O., Smith, C. A., & Thornberry, T. P. (2002). Developmental issues in the impact of child maltreatment on later delinquency and drug use. Criminology, 40, 359 - 399.
Kendall-Tackett, K. A., & Eckenrode, J. (1996). The effects of neglect on academic achievement and disciplinary problems: Developmental perspective. Child Abuse and Neglect, 20, 161 - 169.
Kinard, E. M. (2001a ). Characteristics of maltreatment experience and academic functioning among maltreated children. Violence and Victims, 16, 323 - 337.
Kinard, E. M. (2001b ). Perceived and actual academic competence in maltreated children. Child Abuse and Neglect, 25, 33 - 45.
Korbin, J. E. (2003). Neighborhood and community connectedness in child maltreatment research. Child Abuse and Neglect, 27, 137 - 140.
Leiter, J., & Johnsen, M. C. (1997). Child maltreatment and school performance declines: An eventhistory analysis. American Educational Research Journal, 34, 563 - 589.
Leiter, J., & Johnsen, M. C. (1994). Child maltreatment and school performance. American Journal of Education, 102, 154 - 189.
Lemmon, J. H. (1999). How child maltreatment affects dimensions of juvenile delinquency in a cohort of low-income urban youths. Justice Quarterly, 16, 357 - 376.
Maughan, A., & Cicchetti, D. (2002). Impact of child maltreatment and interadult violence on children's emotion regulation abilities and socioemotional adjustment. Child Development, 73, 1525 - 1542.
Polansky, N., Gaudin, A., Ammons, P., & Davis, K. (1985 ). The psychological ecology of the neglectful mother. Child Abuse & Neglect, 9, 265 - 275.
Rankin, B. H., & Quane, J. M. (2002). Social contexts and urban adolescent outcomes: The interrelated effects of neighborhoods, families, and peers on African-American youth. Social Problems, 49, 79 - 100.
Rogers, W. (1993). Calculation of quantile regression standard errors. Stata Technical Bulletin, 13, 18 - 19.
Sampson, R. J., Morenoff, J. D., & Gannon-Rowley, T. (2002). Assessing neighborhood effects: Social processes and new directions in research. Annual Review of Sociology, 28, 443 - 478.
Sampson, R. J., Raudenbush, S. W., & Earls, F. (1997). Neighborhoods and violent crime: A multilevel study of collective efficacy. Science, 277, 918 - 924.
Sedlak, A. J., & Broadhurst, D. D. (1996). Third national incidence study of child abuse and neglect: Final report. Washington, DC: U.S. Department of Health and Human Services.
Shaw, C., & McKay, H. (1942). Juvenile delinquency and urban areas. Chicago: University of Chicago Press.
Stouthamer-Loeber, M., Wei, E. H., Homish, D. L., & Loeber, R. (2002). Which family and demographic factors are related to both maltreatment and persistent serious juvenile delinquency? Children's Services: Social Policy, Research, and Practice, 5, 261 - 272.
Thornberry, T. P., Ireland, T. O., & Smith, C. A. (2001). The importance of timing: The varying impact of childhood and adolescent maltreatment on multiple problem outcomes. Development and Psychopathology, 13, 957 - 979.
Veltman, M. W. M., & Browne, K. D. (2001). Three decades of child maltreatment research: Implications for the school years. Trauma, Violence, and Abuse, 2, 215 - 239.
Vinson, T., Baldry, E., & Hargreaves, J. (1996). Neighbourhoods, networks, and child abuse. British Journal of Social Work, 26, 523 - 543.
Wikstrom, P., & Loeber, R. (2000 ) Do disadvantaged neighborhoods cause well-adjusted children to become adolescent delinquents? A study of male serious offending, individual risk and protective factors and neighborhood context. Criminology, 38, 1109 - 1142.
Zielinski, D. S., & Bradshaw, C. P. (2006 ). Ecological influences on the sequelae of child maltreatment: A review of the literature. Child Maltreatment, 11, 49 - 62.
Zingraff, M. T., Leiter, J., Myers, K. A., & Johnsen, M. C. (1993). Child maltreatment and youthful problem behavior. Criminology, 31, 173 - 202.
Constance L. Chapple, PhD
University of Oklahoma
Jamie Vaske, PhD
Western Carolina University
Acknowledgments. We would like to thank our reviewers for their helpful comments on revising this manuscript.
Correspondence regarding this article should be directed to Constance L. Chapple, PhD, The University of Oklahoma, Department of Sociology, 780 Van Vleet Oval, Kaufman Hall 331, Norman, OK 73019-2033. E-mail: email@example.com…
Questia, a part of Gale, Cengage Learning. www.questia.com
Publication information: Article title: Child Neglect, Social Context, and Educational Outcomes: Examining the Moderating Effects of School and Neighborhood Context. Contributors: Chapple, Constance L. - Author, Vaske, Jamie - Author. Journal title: Violence and Victims. Volume: 25. Issue: 4 Publication date: July 1, 2010. Page number: 470+. © Springer Publishing Company 2009. Provided by ProQuest LLC. All Rights Reserved.
This material is protected by copyright and, with the exception of fair use, may not be further copied, distributed or transmitted in any form or by any means.